Evaluation Briefs
         Analyzing Quantitative Data for                                                                      No. 20  |  updated August 2018
         Evaluation
         This brief focuses on analyzing quantitative data that your program has collected. It includes an overview of quantitative 
         data; planning quantitative data analysis; conducting quantitative data analysis; and advantages and disadvantages of using 
         quantitative data.
         Overview                                                             Planning quantitative data analysis
         Quantitative data are information in numeric form. They              Quantitative data collection can be difficult and time- 
         can either be counted (such as the number of people who              consuming. It is important to plan your analysis before 
         attend a training) or compared on a numerical scale (such            you collect your data to ensure that your data will be 
         as the number of training participants who said that a               meaningful and useful.
         training was “very helpful” or “somewhat helpful”).                  Determine your focus. Consider the purpose of your 
         Indicators for School Health Programs results are an                 evaluation. Each piece of information you collect should be 
         example of quantitative data that your program collects              useful in understanding and improving your program. Your 
         annually.                                                            data analysis will provide the information that you need to 
         There are two main types of quantitative data:                       do so. Decide how you will use your data to improve your 
                                                                              program before you collect them.
         •  Categorical data have a limited number of possible                Decide who will analyze the data. Your data analyst 
            values. For some categorical data, numbers assigned               should have training and experience in the analysis 
            to categories have no inherent meaning and the order              procedures and software used. When more than one 
            of the categories is arbitrary. For example, when asking 
            about marital status, there are a limited set of possible         person analyzes your data, everyone must use the same 
            responses and categories can be ordered in numerous               systematic approach.
            ways. For other kinds of categorical data, numbers                Develop a data management system. If it is not already 
            assigned to categories have inherent meaning and the              in place, you will need to develop a data management 
            order of the categories follows a logical progression in 
            the values assigned to responses. A question where                system to store and organize your data, such as 
            the responses range from 1 = “strongly agree” to 5                spreadsheets or databases. This system will help to 
            = “strongly disagree” is an example of this type of               improve the quality of data entry and management. Often, 
            categorical data. There is no set interval between each           you can export data directly from your data management 
            response for categorical data.                                    system into quantitative data analysis software.
         •  Continuous data, in contrast, have many possible                  Clean your data. It is likely that there are occasional 
            values. There is a logical progression in the numerical           errors in your data. For example, some fields may have 
            values assigned to responses and the interval between 
            values is meaningful. Continuous data can have almost             been unintentionally left blank. Once your data have been 
            any numeric value along a continuum and can be                    entered into your data management system, review them 
            broken down into smaller parts and still have meaning.            for errors and make adjustments as needed. More than 
            Age, weight, height, and income are all examples of               one person should clean the data to ensure they are error- 
            continuous data.                                                  free.
         Quantitative data analysis is the process of using statistical 
         methods to describe, summarize, and compare data. Your               Obtain data analysis software. There are many popular 
         analysis will vary based on the type of data you collect (see        computer programs that can be used to analyze your 
         below). Analyzing quantitative data allows your evaluation           quantitative data. For the basic statistical methods 
         findings to be more understandable so you can use them               described in this brief, you can use spreadsheets or 
         to strengthen your program.                                          database programs. For more advanced statistics, you 
                                                                              can use a statistical software package. Your data analyst 
                                                                              should be familiar with the software package you choose.
                                                                                                                   C296013-P  November 19, 2018
                                                                              Communicate your findings. When your analysis is 
         Conducting quantitative data analysis
         There are three major steps to this process:                         complete, share your data with stakeholders. There are 
                                                                              several ways to disseminate your findings, including print 
         Conduct statistical tests. You will likely use basic                 formats, oral presentations, and web-based distribution 
         descriptive statistics to explore the main characteristics of        (see Evaluation Brief 9: Disseminating Programs 
         your data. Commonly used statistics include the following:           Achievements and Evaluation Findings to Garner Support).
         •  Frequencies, or counts, describe how many times                   Advantages of using quantitative data
            something has occurred within a given interval, such 
            as a particular category or period of time. For example,          •  Common types of analysis are relatively quick and easy.
            the number of training participants who are classroom 
            teachers is a frequency. Frequencies can be used for              •  Answers the “what” and “how many” questions of 
            categorical or continuous data.                                      evaluation activities.
         •  A percentage is the given number of units divided                 •  Findings are concrete with minimal possibility for 
            by the total number of units and multiplied by 100.                  reviewer bias.
            Percentages are a good way to compare two different 
            groups or time periods. For example, if 50 of 100 training        Disadvantages of using quantitative data
            participants are classroom teachers, 50% of training 
            participants are classroom teachers. Percentages can              •  Data collection can be time-consuming.
            be used for categorical or continuous data.                       •  May not answer the “why” of evaluation activities.
         •  A ratio shows the numerical relationship between two              •  For more advanced data analysis, software and training 
            groups. For example, the ratio of the number of students             needed for analysis can be costly.
            in a particular school (300) to the number of teachers in 
            that same school (25) would be 300/25, or 12:1. Ratios            Resources
            can only be used for continuous data.
         •  Mean, median, and mode are three measures of                      Clayton, R.R. & Crosby, R.A. (2006). Measurement in health 
            the most typical values in your dataset (also called              promotion. In: R.A. Crosby & R.J. DiClemente. Research 
            measures of central tendency). A mean, or average, is             Methods in Health Promotion (229-259). California: Jossey 
                                                                              Bass.
            determined by summing all the values and dividing by              Evaluation Brief 9: Disseminating Program Achievements and 
            the total number of units in the sample. A median is              Evaluation Findings to Garner Support. 
            the 50th percentile point, with half of the values above 
            the median and half of the values below the median.               Available at https://www.cdc.gov/healthyyouth/evaluation/pdf/
            A mode is the category or value that occurs most                  brief9.pdf
            frequently within a dataset.                                      Evaluation Brief 12: Using Graphs and Charts to Illustrate 
                                                                              Quantitative Data. 
         Review and interpret your data. Following data analysis,             Available at https://www.cdc.gov/healthyyouth/evaluation/pdf/
         review your findings to identify patterns in your data.              brief12.pdf
         Consider similarities and differences between responses              Evaluation Brief 20. Analyzing Quantitative Data for Evaluation 
         from participants with different characteristics. Determine          Available at: https://www.cdc.gov/healthyyouth/evaluation/
         whether there are any extreme data that fall significantly           pdf/brief20.pdf
         above or below the mean, median, or mode. Those                      Taylor-Powell, E. Program Development and Evaluation: 
         extreme data points may alter some statistics, such as the           Analyzing Quantitative Data. University of Wisconsin- 
         mean.                                                                Extension; 1996. 
         Summarize your data. Develop tables, graphs and charts               Available at https://learningstore.uwex.edu/Assets/pdfs/
         to summarize your data findings (see Evaluation Brief 12:            G3658-06.pdf
         Using Graphs and Charts to Illustrate Quantitative Data). 
         One common way to summarize data findings is a cross-
         tabulation table. These tables consist of rows displaying 
         values for one variable of interest and columns displaying 
         values for another variable of interest. Cross- tabulation 
         tables can compare several groups or time periods at 
         once. You can use these tables to illustrate any of the 
         statistical methods discussed above.
                                                                              For further information or assistance, contact the 
                                                                              Evaluation Research Team at ert@cdc.gov. You can 
                                                                              also contact us via our website: http://www.cdc.gov/
                                                                              healthyyouth/evaluation/index.htm.